1. Field of the Invention
The present invention relates to an apparatus and a method for automatically diagnosing any abnormality such as fault or trouble in a processing equipment and outputting an adequate instruction which contributes to a proper corrective action of a user.
2. Description of the Related Art
It is observed of late that automation is in rapid progress with regard to processing equipments where production of desired articles is performed by successively executing a series of sequences which comprise, for example, execution of a predetermined process, carriage of the articles to a next processing section, and execution of a preset step therein, whereby the processing steps in serial production and so forth are executed substantially in a full automatic mode according to a program prepared in a controller for controlling the processing equipment.
The controller employed in such processing equipment is called PLC (Programmable Logic Controller), where a program for controlling a series of processing steps is incorporated. This program is prepared by a user in a ladder language which is different from any ordinary language used normally in computers.
Since this program incorporated in a PLC is intended to be used for controlling a sequence, the program can be so prepared that, when any abnormality such as fault has occurred in a processing equipment, an alarm is generated to indicate the fact of occurrence of such fault as a portion of the sequence, but it is difficult to include, in the program, an abnormality diagnosis which instructs the user to perform a required action against the faulty portion or the fault itself. Further, any fault capable of generating an alarm needs to be predictable in the processing equipment, and no alarm can be generated at all with regard to any nonpredicted abnormality.
Therefore the user himself needs to perform a fault diagnosis but, since the latest processing equipment is complicated, it is difficult even for a skilled person to achieve an exact fault diagnosis. For this reason, there is currently adopted a method which carries out fault diagnosis in a processing equipment by the use of a stand-alone computer.
In this case, a so-called expert system is frequently employed as software for fault diagnosis. This expert system relates to one of artificial intelligence application fields and, by previously incorporating knowledge and know-how of solution in a computer, assists services at a level equivalent to or higher than that attained by a skilled person or a human expert.
In the use of this expert system for fault diagnosis, knowledge required for such diagnosis is inputted previously in a computer. And in practical fault diagnosis, the computer successively asks the user some questions with regard to the fault. Then the user successively inputs into the computer the answers to the questions. Subsequently the computer diagnoses the fault from the information inputted by the user and on the basis of the knowledge and know-how (knowledge data base) stored in the computer itself, thereby finding a means of solution and outputting the resultant information to the user.
As mentioned above, the conventional abnormality diagnosis system known heretofore is based on the premise that the knowledge required for the diagnosis needs to be previously inputted into the computer, and the user should answer the questions outputted from the computer employing the expert system. Consequently, it is necessary for the user to be well acquainted with the construction of the processing equipment and the sequence, hence raising some problems that the relevant system is not exactly handleable by any ordinary person and the period of time from occurrence of a trouble to a halt of the processing equipment is rendered long.
In view of the above circumstances, the present applicant proposed, in Japanese Patent Application No. Hei 5-219122, an abnormality diagnosis system wherein a computer is connected to a PLC for controlling the operation of a processing equipment, then any abnormal state in the processing equipment is detected automatically by the computer, and the result relative to the detected abnormal state is displayed.
The computer in such proposed abnormality diagnosis system analyzes a program acquired from the PLC while acquiring input/output data relative to the operation sequence of the processing equipment, then detects any abnormality in the processing equipment from the acquired data and the result of the analysis in accordance with the knowledge data base of an expert system and so forth provided in advance, and supplies to the user the information indicative of the detected abnormal state.
However, the abnormality diagnosis system in the related art mentioned above is the one employing an expert system, and it is based on the premise that the expert system needs to be previously informed of the structure of the program in the PLC, whereby the general-purpose usability thereof is insufficient.